Eur J Clin Pharmacol (2007) 63:677–686
`DOI 10.1007/s00228-007-0312-6
`
`PHARMACOKINETICS AND DISPOSITION
`
`The influence of hepatic impairment
`on the pharmacokinetics of the dipeptidyl peptidase IV
`(DPP-4) inhibitor vildagliptin
`
`Y.-L. He & R. Sabo & J. Campestrini & Y. Wang &
`M. Ligueros-Saylan & K. C. Lasseter & S. C. Dilzer &
`D. Howard & W. P. Dole
`
`Received: 8 January 2007 / Accepted: 4 April 2007 / Published online: 8 May 2007
`# Springer-Verlag 2007
`
`Abstract
`Objective Vildagliptin is a potent and selective dipeptidyl
`peptidase-IV (DPP-4) inhibitor that
`improves glycemic
`control
`in patients with type 2 diabetes mellitus by
`increasing α- and β-cell responsiveness to glucose. This
`study investigated the pharmacokinetics of vildagliptin in
`patients with hepatic impairment compared with healthy
`subjects.
`Methods This was an open-label, parallel-group study in
`patients with mild (n=6), moderate (n=6) or severe (n=4)
`hepatic impairment and healthy subjects (n=6). All subjects
`received a single 100-mg oral dose of vildagliptin, and
`plasma concentrations of vildagliptin and its main pharma-
`cologically inactive metabolite LAY151 were measured up
`to 36 h post-dose.
`Results Exposure to vildagliptin (AUC0–∞ and Cmax)
`decreased non-significantly by 20 and 30%, respectively,
`in patients with mild hepatic impairment [geometric mean
`
`Y.-L. He (*) : W. P. Dole
`Exploratory Development,
`Novartis Institutes for Biomedical Research, Inc.,
`400 Technology Square, Building 605, Rm 819,
`Cambridge, MA 02139-3584, USA
`e-mail: yanling.he@novartis.com
`R. Sabo : Y. Wang : M. Ligueros-Saylan : D. Howard
`Novartis Pharmaceuticals Corporation,
`East Hanover, NJ, USA
`
`J. Campestrini
`Novartis Pharma SA,
`Rueil-Malmaison, France
`K. C. Lasseter : S. C. Dilzer
`Pharmanet Development Group,
`Miami, FL, USA
`
`ratio (90% CI): AUC0–∞, 0.80 (0.60, 1.06), p=0.192; Cmax,
`0.70 (0.46, 1.05), p=0.149]. Exposure to vildagliptin was
`also decreased non-significantly in patients with moderate
`hepatic impairment [−8% for AUC0–∞, geometric mean ratio
`(90% CI): 0.92 (0.69, 1.23), p=0.630; −23% for Cmax,
`geometric mean ratio (90% CI): 0.77 (0.51, 1.17), p=0.293].
`In patients with severe hepatic impairment, Cmax was 6%
`lower than that in healthy subjects [geometric mean ratio
`(90% CI): 0.94 (0.59, 1.49), p=0.285], whereas AUC0–∞
`was increased by 22% [geometric mean ratio (90% CI): 1.22
`(0.89, 1.68), p=0.816). Across the hepatic impairment
`groups, LAY151 AUC0–∞ and Cmax were increased by 29–
`84% and 24–63%, respectively, compared with healthy
`subjects. The single 100-mg oral dose of vildagliptin was
`well tolerated by patients with hepatic impairment.
`Conclusions There was no significant difference in expo-
`sure to vildagliptin in patients with mild, moderate or
`severe hepatic impairment; therefore, no dose adjustment
`of vildagliptin is necessary in patients with hepatic
`impairment.
`
`Keywords Chronic liver disease . Dipeptidyl
`peptidase-IV inhibitor . Pharmacokinetics . Type 2 diabetes .
`Vildagliptin
`
`Introduction
`
`The incretin hormones, glucagon-like peptide-1 (GLP-1)
`and glucose-dependent insulinotropic polypeptide (GIP) are
`released from the gastrointestinal tract into the portal vein
`in response to food intake [1, 11]. GLP-1 and GIP are
`required for the maintenance of normal glucose tolerance,
`and they stimulate insulin release in a glucose-dependent
`
`Mylan EX 1013, Page 1
`
`

`

`678
`
`Eur J Clin Pharmacol (2007) 63:677–686
`
`manner [15, 21]. In healthy subjects, GLP-1 and GIP
`activity may be responsible for up to 70% of the insulin
`secreted in response to a meal [17]. In addition, GLP-1 also
`acts to suppress glucagon release [10]. Patients with type 2
`diabetes mellitus show impaired GLP-1 secretion, which
`may explain, at least in part, the reduced incretin effect seen
`in these patients [17, 22]. As GLP-1 is rapidly inactivated
`by the serine peptidase dipeptidyl peptidase IV (DPP-4) [5],
`a novel therapeutic approach to the treatment of type 2
`diabetes is to inhibit DPP-4 activity, and thereby prolong
`the physiological actions of GLP-1 [6].
`Vildagliptin (LAF237) is a potent, selective, orally active
`DPP-4 inhibitor developed for the treatment of type 2
`diabetes mellitus. In a clinical trial in patients with type 2
`diabetes, vildagliptin 100 mg once daily for 4 weeks
`increased post-prandial active GLP-1, reduced glucagon
`levels and improved glycemic control compared with
`placebo [3]. A recent study in drug-naive patients with type
`2 diabetes has also shown that vildagliptin improves
`pancreatic β-cell function by increasing the rate of insulin
`secretion [13]. Moreover, in a long-term study, the addition
`of vildagliptin 100 mg once daily to metformin therapy
`significantly improved glycemic control in patients with type
`2 diabetes mellitus over the 12 months of treatment [2].
`the
`Chronic liver disease can significantly impair
`function of hepatic drug-metabolizing enzymes (particular-
`ly microsomal oxidases), reduce hepatic blood flow and
`alter renal function [16, 19, 24]. The main metabolic
`pathway for vildagliptin in humans is hydrolysis to the
`inactive metabolite LAY151, which accounts for approxi-
`mately 69% of total elimination of an oral dose. Although
`85% of an oral dose of vildagliptin is ultimately excreted by
`the kidney, the liver is one of the major sites of vildagliptin
`metabolism (Novartis, data on file). Liver, kidney and
`intestinal microsomes are all capable of hydrolyzing
`vildagliptin to LAY151 in vitro. The aim of this study was
`to assess the effect of hepatic impairment on the disposition
`of vildagliptin by evaluating the pharmacokinetics, safety
`and tolerability of a single 100-mg oral dose of vildagliptin
`in patients with mild, moderate or severe hepatic impairment
`compared with healthy subjects with normal liver function.
`
`Methods
`
`Study design
`
`This was an open-label, single-dose, parallel-group study
`conducted at a single study center. Following a 21-day
`screening period, eligible subjects underwent baseline
`evaluation (Day −1) and received a single 100-mg oral dose
`of vildagliptin on Day 1. Blood samples were taken at
`regular intervals after dosing for the measurement of plasma
`
`drug levels. End-of-study evaluations were performed
`immediately after the last blood sample (36 h post-dose).
`Selection of the 100-mg dose of vildagliptin was based on
`the highest expected clinical dose. Treatment compliance
`was assured by the administration of vildagliptin under
`supervision of study center personnel and confirmed by the
`presence of vildagliptin in the plasma and urine.
`All subjects were admitted to the study center at least
`14 h prior to vildagliptin administration, and they were
`discharged following completion of the end-of-study evalua-
`tion. Vildagliptin was administered as a single tablet with 240
`mL of water between 7:00 and 8:30 A.M. following an
`overnight fast of at least 10 h. All subjects continued fasting
`for another 2 h after dosing, and unless performing a study
`assessment, subjects were required to rest quietly for a further
`4 h.
`The primary objective of this study was to assess the
`single-dose pharmacokinetics of vildagliptin 100 mg in
`patients with stable chronic liver disease compared with
`healthy subjects with normal liver function. A secondary
`objective was to assess the safety and tolerability of
`vildagliptin in patients with impaired hepatic function.
`
`Study population
`
`The study recruited men or women (ages: 18–60 years) of
`body weight ≥55 kg with a body mass index (BMI) of 22–
`40 kg/m2, resting pulse rate of 60–100 beats per minute and
`platelet count ≥50×109/L at screening and baseline. Female
`subjects had to be surgically sterile, postmenopausal, or use
`a double-barrier method of contraception.
`A total of 22 subjects were recruited, comprising healthy
`control subjects (n=6) and patients with mild (n=6),
`moderate (n=6) or severe (n=4) hepatic impairment. The
`diagnosis of cirrhosis was based on medical history and the
`presence of physical signs (liver firmness to palpitation,
`splenic enlargement, spider angioma, palmar erythema,
`parotid hypertrophy, testicular atrophy or gynecomastia).
`The Child-Pugh clinical assessment score was used to
`evaluate the degree of hepatic impairment based on
`guidance for clinical trials in patients with impaired hepatic
`function from the United States Food and Drug Adminis-
`tration [7]. Mild, moderate and severe hepatic impairment
`were defined by Child−Pugh assessment scores of 5–6, 7–9
`or 10–12, respectively (Table 1) [18].
`Apart from hepatic impairment, all participants were in
`good health, as determined by medical history, physical
`examination, vital signs, electrocardiogram (ECG) and
`laboratory tests at screening. Blood pressure (BP) inclusion
`criteria for healthy volunteers were a systolic BP (SBP) of
`90–160 mmHg and a diastolic BP (DBP) of 60–100 mmHg.
`For patients with hepatic impairment, inclusion criteria were
`an SBP of 90–180 mmHg and a DBP of 60–115 mmHg.
`
`Mylan EX 1013, Page 2
`
`

`

`Eur J Clin Pharmacol (2007) 63:677–686
`
`679
`
`Table 1 Child-Pugh classification scores for patients with mild,
`moderate or severe hepatic disease
`
`Mild hepatic
`impairment
`(n=6)
`
`Moderate
`hepatic
`impairment
`(n=6)
`
`Severe
`hepatic
`impairment
`(n=4)
`
`Overall
`Encephalopathy
`Ascites
`Bilirubin
`Albumin
`Prothrombin
`time
`
`6
`1–2
`1–2
`1
`1
`1
`
`7–8
`2
`2
`1–2
`1–2
`1
`
`10–12
`2
`1–3
`1–3
`2–3
`1–3
`
`Data are shown as minimum-maximum values for overall score and
`each component score
`
`Exclusion criteria for healthy volunteers included subjects
`who had used any prescription medication within 1 month
`prior to dosing or over-the-counter medications or vitamins
`within 14 days prior to dosing. Additional exclusion criteria
`for patients with hepatic impairment included: symptoms or
`history of Stage II (or worse) degree of encephalopathy within
`6 months of study entry; clinical evidence of severe ascites;
`history of surgical portosystemic shunt; prothrombin time >18
`s; any evidence of progressive liver disease within the
`previous 4 weeks, as
`indicated by changes in hepatic
`transaminases, alkaline phosphatase and γ-glutamyltransfer-
`ase, or a ≥50% worsening of serum bilirubin or prothrombin
`time. Patients with hepatic impairment were also excluded if,
`in the opinion of the study investigator,
`the degree of
`encephalopathy impaired the ability to provide written
`informed consent.
`Study participants were not permitted to engage in
`strenuous physical exercise for 7 days before dosing or to
`take alcohol for 72 h before dosing until after the study
`completion evaluation. Intake of xanthine-containing foods or
`beverages was discontinued 48 h before dosing and not
`permitted while participants were admitted to the study center.
`The study protocol was approved by the relevant local
`ethical
`review board. This study was conducted in
`accordance with the Guidelines for Good Clinical Practice
`and adhered to the ethical principles of the Declaration of
`Helsinki. All subjects provided written informed consent.
`
`Pharmacokinetic measurements
`
`Blood samples were taken by either direct venipuncture or
`by an indwelling cannula inserted in a forearm vein, at 0
`(pre-dose), 0.5, 1, 1.5, 2, 2.5, 3, 4, 6, 8, 10, 12, 15, 24 and
`36 h after vildagliptin administration. Blood samples were
`collected into sodium or lithium heparin tubes and stored at
`−70°C prior to analysis. Plasma samples were split equally
`for analysis of vildagliptin and LAY151. Urine samples
`
`were collected during the following time periods: pre-dose
`(overnight), 0–12, 12–24 and 24–36 h post-dose; urine
`collected during each time period was pooled, and a 5-mL
`aliquot was frozen.
`Plasma and urinary concentrations of vildagliptin and
`plasma concentrations of LAY151 were determined by liquid
`chromatography/mass spectrometry/mass spectrometry (LC/
`MS/MS) methods. All sample extractions were performed on
`10-mg Oasis HLB 96-well plates (Waters, Milford, Mass.)
`using an Automated Liquid Handling System model Multi
`probe II Plus HT EX. For the measurement of vildagliptin in
`plasma or urine, samples and internal standards were diluted
`in 4 mmol/L ammonium acetate (urine samples were also
`mixed with water). Samples were transferred to the
`extraction plates and successively washed with methanol:2%
`ammonium hydroxide (5:95, v/v), methanol:2% ammonium
`hydroxide (20:80, v/v) and water, and then eluted twice with
`methanol:0.1% trifluoroacetic acid (TFA; 80:20, v/v). Sam-
`ples were evaporated to a final volume of approximately
`50 μL and then diluted with methanol:0.5% ammonium
`hydroxide (15:85, v/v). For the measurement of LAY151 in
`plasma, samples and internal standards were mixed with
`4 mmol/L ammonium acetate. Samples were transferred to
`the extraction plates, successively washed with 0.1% TFA
`and 0.01% TFA and then eluted twice with methanol:0.5%
`ammonium acetate (80:20, v/v).
`The samples were analyzed by high-performance liquid
`chromatography (HPLC) using an XTerra MS C18 5 μm
`(150×2.1 mm) column (Waters) at 30°C with isocratic
`elution using 40% mobile phase A [10 mM ammonium
`acetate-methanol (95:5, v/v), pH 8] and 60% mobile phase B
`[acetonitrile-methanol (10/90, v/v)] at a flow rate of 0.2 mL/
`min. Detection was performed by MS/MS using an API3000
`(Applied Biosystems, Foster City, Calif.) or Quantum
`Discovery (Thermo Finnigan, San José, Calif.) mass
`spectrometer. The general conditions used were: positive
`ion mode with turbo ion spray or electro spray ionization.
`The masses for vildagliptin were precursor ion m/z 304 and
`product ion m/z 154; the masses for LAY151 were precursor
`m/z 323 and product m/z 173. For vildagliptin, the lower limit
`of quantification (LOQ) was 2 ng/mL in 0.2 mL of plasma and
`5 ng/mL in 0.1 mL of urine. The internal standard for
`vildagliptin was (2S)-1-[[(3-hydroxytricyclo[3.3.1.13,7]dec-
`1-yl)amino]acetyl]-2-pyrrolidine-2,3,4,5-13C-15N-carbonitrile-
`cyano-13C (Novartis Pharmaceuticals Corp, N.J.). Within-study
`assay validation showed an assay precision (co-efficient of
`variation) of 3.1–6.0% (bias of −4.2 to 3.5%) for plasma
`samples (nominal concentrations of 5.25, 400 and 900 ng/mL),
`and a precision of 6.5–7.4% (bias of −5.5 to 1.3%) for urine
`samples (nominal concentrations of 15, 2000 and 4000 ng/
`mL). For LAY151, the LOQ was 2 ng/mL in 0.2 mL of
`plasma. The internal standard was [D7] LAY151 (Novartis
`Pharma AG, Basel, Switzerland), and within-study assay
`
`Mylan EX 1013, Page 3
`
`

`

`680
`
`Eur J Clin Pharmacol (2007) 63:677–686
`
`validation showed a precision of 2.5–3.6% and a bias of −2.4
`to 5.9% (nominal concentrations of 5.01, 5.03, 406, 412, 902
`and 915 ng/mL). A single laboratory assayed all plasma and
`urine samples. Within-study assay validation was performed
`by an analysis of quality control samples.
`
`Safety and tolerability assessments
`
`Safety and tolerability assessments included the monitoring
`and recording of all adverse events (AEs), concomitant
`medications and significant non-drug therapies. Assessment
`of standard blood chemistries, hematological profile,
`urinalysis, physical examination, ECG and vital signs were
`performed at screening, baseline and at the end of the study.
`
`ranged from 0.25 to 0.29 for AUC and Cmax, respectively.
`Statistical comparisons of pharmacokinetic parameters be-
`tween the hepatic impairment groups and the healthy
`subjects group were performed for
`log-transformed
`AUC0–∞, AUC0–36 h and Cmax of vildagliptin using an
`analysis of variance (ANOVA) model. The source of
`variation included in the ANOVA model was hepatic group.
`Geometric mean ratios and corresponding 90% confidence
`intervals (CI) for each hepatic impairment group versus the
`healthy subjects group were calculated. Statistical compar-
`isons of pharmacokinetic parameters were performed using
`the PROC MIXED SAS procedure (SAS Institute, Cary, N.
`C.).
`
`Pharmacokinetic analyses
`
`Results
`
`The following pharmacokinetic parameters were calculated
`for vildagliptin using plasma concentration data: Cmax
`(maximum plasma concentration), tmax (time to reach Cmax)),
`AUC0–36 h (area under the plasma concentration–time curve
`for time period 0–36 h), AUC0–∞ (area under the plasma
`concentration-time curve extrapolated from time 0 to
`infinity),
`t1=2
`(elimination half-life associated with the
`terminal rate constant), CL/F (total clearance of drug from
`plasma, corrected for bioavailability F), Vz/F (apparent
`volume of distribution, corrected for bioavailability F) and
`CLR (renal clearance of drug from plasma). The total amount
`of vildagliptin excreted into urine was also measured. CLR
`was calculated using both plasma and urine data over the
`36-h assessment period. The following pharmacokinetic
`parameters were calculated for LAY151 using plasma
`concentration data: Cmax, tmax, AUC0–t, AUC0–∞ and t1=2
`.
`Pharmacokinetic parameters were obtained according to
`the non-compartmental methods using WINNONLIN PRO
`(ver. 4.0; Pharsight Corp., Mountain View, Calif.).
`
`Statistical analyses
`
`The sample size was determined mainly on the basis of the
`sensitivity of statistical testing to detect a clinically meaning-
`ful change in drug exposure between patients with hepatic
`impairment and healthy volunteers, and with reference to the
`sample sizes commonly used in pharmacokinetic studies in
`patients with hepatic impairment. Assuming an inter-subject
`co-efficient of variation (CV) of 0.25, a sample size of six in
`each of the three hepatic impairment patient subgroups and in
`the healthy subject group would ensure an 80% power to
`detect a 45% difference in pharmacokinetic parameter values
`between the healthy and hepatic-impaired groups, based on a
`two-sample, two-sided t-test at a 95% significance level.
`Previous vildagliptin pharmacokinetic studies in patients
`with type 2 diabetes mellitus showed that inter-subject CV
`
`Patient characteristics
`
`A total of 22 subjects were enrolled and completed the study
`(i.e. received vildagliptin and completed all post-dosing
`evaluations). Although the study originally planned to
`recruit six subjects in each group; only four subjects were
`enrolled in the severe hepatic impairment group due to the
`difficulty of recruiting these patients. Several patients were
`recruited with minor protocol deviations; one subject had a
`BMI (18.5 kg/m2) below the lower limit defined in the
`inclusion criteria; six participants were older than the
`maximum specified age of 60 years; one subject had a
`platelet count that was below that stated in the inclusion
`criteria. However, none of the deviations were considered
`significant by the investigator, and these subjects were
`allowed to participate in this study.
`Patients with hepatic impairment and healthy subjects
`showed similar baseline and demographic characteristics
`(Table 2), except that mean age was lower in the healthy
`subjects group (37.8 years) than in the hepatic impairment
`subgroups (54.8–56.5 years). The majority of patients with
`hepatic impairment had a medical history of alcoholism,
`which was not an active condition at the time of the study,
`or had existing medical conditions related to hepatic
`impairment (Table 2). Prothrombin time was increased in
`patients with mild, moderate or severe hepatic impairment
`compared with healthy subjects and was the longest in the
`severe hepatic impairment group. Total bilirubin was also
`markedly higher in patients with severe hepatic impairment
`than in patients in the other groups. Overall, ten (62.5%)
`patients with hepatic impairment were receiving medication
`for the treatment of hepatic disease or associated illnesses;
`the most common co-medication was spironolactone. There
`were no significant past or present medical conditions in the
`healthy volunteers group, and none of the healthy subjects
`were taking medications prior to enrolment.
`
`Mylan EX 1013, Page 4
`
`

`

`Eur J Clin Pharmacol (2007) 63:677–686
`
`Table 2 Baseline and demographic characteristics
`
`681
`
`Characteristic
`
`Healthy subjects (n=6)
`
`Patients with hepatic impairment
`
`Mild (n=6)
`
`Moderate (n=6)
`
`Severe (n=4)
`
`Age (years)
`Range
`Weight (kg)
`Height (cm)
`Gender, n (%)
`Male
`Female
`Ethnicity, n (%)
`Caucasian
`Black
`Othera
`Etiology of hepatic disease, n (%)
`Alcohol-induced liver cirrhosis
`Hepatitis C
`Clinical manifestations of hepatic disease
`Encephalopathy, n (%)
`Ascites, n (%)
`Peripheral edema, n (%)
`Albumin (g/dL)
`Total bilirubin (mg/dL)
`Prothrombin timeb (s)
`Creatinine clearancec (mL/min/1.73m2)
`
`37.8±10.9
`19–53
`70.3±9.4
`165±7
`
`3 (50)
`3 (50)
`
`0
`0
`6 (100)
`
`0
`0
`
`0
`0
`0
`4.4±0.1
`0.9±0.3
`11.8±0.4
`81.6±7.0
`
`56.0±5.8
`49–64
`79.2±16.2
`168±9
`
`3 (50)
`3 (50)
`
`4 (66.7)
`1 (16.7)
`1 (16.7)
`
`6 (100)
`3 (50)
`
`5 (83.3)
`4 (66.7)
`2 (33.3)
`4.3±0.2
`0.7±0.3
`12.1±0.8
`78.2±13.4
`
`56.5±6.6
`48–63
`80.5±17.8
`171±13
`
`5 (83)
`1 (17)
`
`1 (16.7)
`4 (66.7)
`1 (16.7)
`
`6 (100)
`3 (50)
`
`6 (100)
`4 (66.7)
`1 (16.7)
`3.8±0.5
`1.0±0.5
`12.2±1.1
`69.1±7.9
`
`54.8±14.7
`44–76
`72.8±20.3
`163±14
`
`2 (50)
`2 (50)
`
`1 (25)
`1 (25)
`2 (50)
`
`3 (75)
`4 (100)
`
`4 (100)
`4 (100)
`1 (25)
`3.0±0.6
`2.4±0.9
`16.1±2.6
`91.1±33.5
`
`Data are presented as mean±SD unless otherwise stated
`a ‘Other’ ethnic category includes ethnically Hispanic subjects of either Caucasian or black heritage
`b Prothrombin time was measured at screening
`c Calculated using the formula [(140–age in years) × weight in kg]/(serum creatinine × 72)
`
`Pharmacokinetics of vildagliptin in patients with mild,
`moderate or severe hepatic impairment
`
`The plasma concentration-time profiles for vildagliptin are
`illustrated in Fig. 1. The pharmacokinetic parameters for both
`vildagliptin and LAY151 are summarized in Table 3.
`Exposure to vildagliptin (AUC and Cmax) in patients with
`mild or moderate hepatic impairment was lower than in
`healthy subjects following the administration of a single 100-
`mg oral dose. In patients with mild hepatic impairment,
`vildagliptin AUC0–∞ and Cmax were reduced by 20% and 30,
`respectively, compared with healthy subjects [geometric mean
`ratio (90% CI): AUC0–∞, 0.80 (0.60, 1.06); Cmax, 0.70 (0.46,
`1.05)]. The reductions in AUC0–∞ and Cmax were not
`statistically significant compared with healthy subjects
`(p=
`0.192 and p=0.149, respectively). Vildagliptin AUC0–∞ was
`reduced by 8% and Cmax by 23% in patients with moderate
`hepatic impairment compared with healthy subjects [mean
`ratio (90% CI): AUC0–∞, 0.92 (0.69, 1.23); Cmax, 0.77 (0.51,
`1.17)]; however,
`these differences were not statistically
`significant (AUC0–∞, p=0.630 and Cmax, p=0.293 vs. healthy
`subjects). Plasma concentrations of vildagliptin following the
`administration of a single 100-mg oral dose were higher in
`patients with severe hepatic impairment than in healthy subjects
`
`Fig. 1 Plasma concentration−time profiles of vildagliptin in healthy
`subjects and patients with mild, moderate or severe hepatic impair-
`ment. Plasma concentrations of vildagliptin are shown after the
`administration of a single 100-mg oral dose to healthy subjects (n=6)
`and patients with mild (n=6), moderate (n=6) or severe (n=4) hepatic
`impairment. Data are presented as mean±SEM
`
`Mylan EX 1013, Page 5
`
`

`

`682
`
`Eur J Clin Pharmacol (2007) 63:677–686
`
`Table 3 Pharmacokinetic parameters of vildagliptin and LAY 151 in healthy subjects and patients with mild, moderate or severe hepatic
`impairment after administration of a single 100-mg oral dose
`
`Parameter
`
`Healthy subjects (n=6)
`
`Patients with hepatic impairment
`
`Vildagliptin
`tmax (h)
`Cmax (ng/mL)
`AUC0–t (ng·h/mL)
`AUC0–∞ ( ng·h/mL)
`t1=
`(h)
`2
`CL/F (L/h)
`Vz/F (L)
`CLR (L/h)
`LAY151
`tmax (h)
`Cmax (ng/mL)
`AUC0–t, (ng·h/mL)
`AUC0–∞ (ng·h/mL)
`t1=
`, h
`
`2
`
`1.3 (1.00–3.00)
`675±263
`2567±428
`2580±425
`2.0±0.5
`39.6±6.3
`115±32
`9.4±4.4
`
`6.0 (6.0–8.0)
`330±99
`5297±1715
`5641±1884
`7.9±0.7
`
`Mild (n=6)
`
`Moderate (n=6)
`
`Severe (n=4)
`
`Severe (n=3)a
`
`1.2 (1.0–2.0)
`497±229
`2076±514
`2101±512
`4.9±4.9
`50.4±14.1
`369±390
`10.2±2.2
`
`8.0 (6.0–10.0)
`406±108
`6721±1675
`7252±1782
`8.4±1.6
`
`1.0 (0.5–3.0)
`512±166
`2411±740
`2437±742
`3.1±1.6
`43.8±11.0
`200±122
`9.0±3.3
`
`9.00 (6.0–12.0)
`483±165
`7902±3168
`8396±3556
`7.2±0.8
`
`2.0 (1.0–4.0)
`632±247
`3322±1472
`3354±1462
`2.4±0.3
`34.4±15.0
`120±55
`6.2±4.0
`
`10.0 (8.0–12.0)
`549±217
`10160±4801
`11545±5964
`9.3±1.5
`
`1.5 (1.0–4.0)
`519±121
`2655±764
`2695±777
`2.4±0.3
`39.6±13.2
`138±50
`8.1±0.9
`
`10.0 (8.0–12.0)
`454±130
`8071±2897
`8893±3338
`8.7±1.0
`
`Data are presented as mean±SD, except for tmax, which is shown as median (minimum–maximum)
`a Omitting one subject who exhibited both hepatic and renal impairment (creatinine clearance: 43.8 mL/min/1.73 m2 )
`
`(Fig. 1, Table 3). Although Cmax and AUC0–∞ were 6% lower
`[geometric mean ratio (90% CI): 0.94 (0.59, 1.49)] and 22%
`higher [1.22 (0.89, 1.68)], respectively, in patients with severe
`hepatic impairment than in healthy subjects, these differences
`were not statistically significant (AUC0–∞, p=0.816; Cmax,
`p=0.285). Exposure to vildagliptin (AUC0–∞ and Cmax) in
`patients with severe hepatic impairment showed a greater
`degree of variability compared with the other groups (Fig. 2).
`Overall, variability in exposure to vildagliptin in patients with
`hepatic impairment appeared to increase with the severity of
`the hepatic impairment (Fig. 2). Changes in exposure were not
`consistent with the severity of hepatic impairment.
`It should be noted that one patient with severe hepatic
`impairment exhibited a vildagliptin AUC0–∞ of 5330 h ng/mL,
`which was considerably higher than the corresponding values
`for the other three patients in this subgroup (1829–3330 h ng/
`mL). Vildagliptin Cmax was also markedly higher in this
`patient
`than in the other patients with severe hepatic
`impairment (971 vs. 384–618 ng/mL, respectively). Moreover,
`CL/F and renal clearance (CLR) in this patient were 18.8 and
`0.33 L/h, respectively compared with values of 30.0–54.7
`(CL/F) and 7.2–9.0 L/h (CLR) for the other patients with
`severe hepatic impairment. This subject also exhibited a
`creatinine clearance of 44.4 mL/min/1.73 m2 compared with
`105.9–145.9 mL/min/1.73 m2 for other patients in the group.
`These observations suggest that this subject may have had an
`impairment of both the liver and the kidneys. Thus, the overall
`pharmacokinetic parameters in the severe hepatic impairment
`subgroup may have been skewed by the data from this patient.
`When these data were omitted from the analysis, exposure to
`vildagliptin in patients with severe hepatic impairment was
`
`actually reduced compared with healthy subjects and consis-
`tent with the trend observed in patients with mild or moderate
`hepatic impairment (Table 3).
`The mean CL/F of vildagliptin was 27% higher in
`patients with mild hepatic impairment
`than in healthy
`subjects, but it was similar in patients with moderate or
`severe hepatic impairment as compared with healthy
`subjects (Table 3). Mean CLR in patients with severe hepatic
`impairment was 34.4% lower than that in healthy subjects;
`however, this difference was minimal when one patient with
`moderate renal impairment (based on calculated creatinine
`clearance) was excluded (Table 3). Median tmax was
`comparable in patients with mild or moderate or severe
`hepatic impairment and healthy subjects, although the range
`for tmax was considerably larger in patients with severe
`hepatic impairment. Mean t1=2
`(4.9 h) appeared to be longer
`in patients with mild hepatic impairment than in healthy
`subjects (2.0 h) and in those with moderate or severe hepatic
`impairment (3.1 h). However, the mild hepatic impairment
`group had a large variability in estimated t1=2
`. with a CV of
`99% due to one subject who had measurable vildagliptin
`plasma concentrations at 36 h post-dose, whereas for all
`other patients, vildagliptin was not detectable in the plasma
`after 24 h.
`
`Pharmacokinetics of LAY151 in patients with mild,
`moderate or severe hepatic impairment
`
`The plasma concentration versus time profiles of LAY151 are
`illustrated in Fig. 3. Mean plasma concentrations of LAY151
`were higher in patients with mild or moderate hepatic
`
`Mylan EX 1013, Page 6
`
`

`

`Eur J Clin Pharmacol (2007) 63:677–686
`
`683
`
`Fig. 3 Plasma concentration–time profiles of LAY151 in healthy
`subjects and patients with mild, moderate or severe hepatic impair-
`ment. Plasma concentrations of LAY151 are shown after
`the
`administration of a single 100-mg oral dose of vildagliptin to healthy
`subjects (n=6) and patients with mild (n=6), moderate (n=6) or
`severe (n=4) hepatic impairment. Data are presented as mean±SEM
`
`impairment (8 and 9 h, respectively) than in healthy subjects
`(6 h; Table 3).
`Compared with healthy subjects, exposure to LAY151 was
`increased to a greater extent in patients with severe hepatic
`impairment than in patients with mild or moderate hepatic
`impairment (Table 3). AUC0–∞ and Cmax were increased by
`63% [geometric mean ratio (90% CI): 1.63 (1.14, 2.08)] and
`94% [mean ratio (90% CI): 1.94 (1.29, 2.91)], respectively,
`compared to healthy subjects. Median tmax was also longer in
`patients with severe hepatic impairment
`than in healthy
`subjects; however, mean t1=
`of LAY151 was comparable
`between the two groups (Table 3). It was notable that
`exposure to LAY151 was considerably higher in one patient
`(AUC0–∞, 19502 h ng/mL; Cmax, 834 ng/mL) than in the
`other
`three patients with severe hepatic impairment
`(AUC0–∞, 7,662–12,672 h ng/mL; Cmax, 411–600 ng/mL).
`This was the same subject who showed moderately impaired
`renal function and exhibited a markedly higher vildagliptin
`exposure than the remainder of the group. The pharmacoki-
`netic parameters of LAY151 in the severe hepatic impair-
`ment subgroup when data from this patient are omitted from
`the analysis are shown in Table 3.
`
`2
`
`Correlation between creatinine clearance and exposure
`to vildagliptin and LAY151
`
`The relationship between creatinine clearance and exposure
`to vildagliptin and LAY151 was assessed because the
`majority of an oral dose of vildagliptin is excreted in the
`
`Fig. 2 Lack of correlation between severity of hepatic impairment
`and (a) Cmax and (b) AUC0–∞ following the administration of a single
`100-mg oral dose of vildagliptin. Figure shows individual patients'
`Child-Pugh score and respective mean for (a) Cmax and (b) AUC0–∞
`for healthy subjects (n=6; filled circles) and patients with mild (n=6;
`open triangles), moderate (n=6; open squares) or severe (n=4; open
`diamonds) hepatic impairment
`
`than in healthy subjects (Table 3). LAY151
`impairment
`AUC0–∞ was 31% and 45% higher in patients with mild or
`moderate hepatic impairment, respectively, compared with
`healthy subjects [geometric mean ratio (90% CI): mild, 1.31
`(0.91, 1.89); moderate, 1.45 (1.01, 2.08)]. Furthermore,
`LAY151 Cmax was increased by 24% [mild hepatic
`impairment, mean ratio (90% CI): 1.24 (0.90, 1.71)] and
`45% [moderate hepatic impairment, mean ratio (90% CI):
`1.45 (1.05, 2.00)] compared with healthy subjects. The
`magnitude of the changes in LAY151 exposure appeared to
`be related to the severity of hepatic impairment. Median tmax
`was also longer in patients with mild or moderate hepatic
`
`Mylan EX 1013, Page 7
`
`

`

`684
`
`Eur J Clin Pharmacol (2007) 63:677–686
`
`urine. There was no significant correlation between creatinine
`clearance and vildagliptin AUC0–∞ (r=−0.246, p =0.270) or
`Cmax (−0.099, p=0.660; Fig. 4a). However, reductions in
`creatinine clearance showed a significant correlation with
`increased LAY151 AUC0–∞ (r=−0.475, p=0.026) and Cmax
`(r=−0.524, p= 0.012; Fig. 4b).
`
`Safety and tolerability
`
`Administration of a single 100-mg oral dose of vildagliptin
`was well tolerated both by patients with hepatic impairment
`
`Fig. 4 Correlation between creatinine clearance (CLcr) and AUC0–∞
`of (a) vildagliptin and (b) LAY151 following administration of a
`single 100-mg oral dose of vildagliptin. Figure shows individual
`patient creatinine clearance values and respective mean AUC0–∞
`values for (a) vildagliptin and (b) LAY151 for healthy subjects (n=6;
`filled circles) and patients with mild (n=6; open triangles), moderate
`(n=6; open squares) or severe (n=4; open diamonds) hepatic
`impairment
`
`and by healthy subjects. Overall, four (18.2%) of the 22
`participants reported a total of five AEs. One AE was
`reported each in the healthy subjects, and mild and
`moderate hepatic impairment groups;
`two AEs were
`reported in the severe hepatic impairment group. The most
`commonly reported AE was somnolence, which was
`reported by one patient each in the mild, moderate and
`severe hepatic impairment subgroups. There was also one
`report of diarrhea each in the healthy subject group and
`severe hepatic impairment group. All reported AEs were
`mild in intensity,
`transient and resolved spontaneously.
`There were no serious AEs.
`No clinically significant abnormalities other than those
`expected in patients with hepatic impairment were detected
`during the assessment of vital signs, blood chemistry,
`hematology or urinalysis.
`Vildagliptin had no notable effects on serum glucose
`levels in healthy volunteers or patients with hepatic
`impairment. In healthy volunteers, the mean serum glucose
`concentration at baselin